Teams need reliable pipelines, clear data models, and repeatable quality checks. This skill provides practical data engineering patterns for batch, streaming, orchestration, and DataOps work.

Computer vision projects often fail when model choices, data quality, and deployment constraints are handled separately. This skill guides Claude, Codex, and Claude Code through practical architecture, training, optimization, and production planning.

Vue teams need consistent state patterns across components, tests, SSR, and Nuxt apps. This skill provides focused Pinia guidance with examples for stores, plugins, hydration, and testing.

Nuxt projects require correct choices across SSR, routing, data fetching, modules, and deployment. This skill gives Claude, Codex, and Claude Code focused Nuxt guidance from framework-oriented references.

Roadmap plans often fail when epics, milestones, and team capacity are estimated separately. This skill helps Claude, Codex, and Claude Code create practical sprint and milestone plans.

Teams often lose alignment when local roadmap files and ClickUp tasks drift apart. This skill provides mapping rules and sync guidance for ClickUp MCP workflows.

End-to-end test runs can overload local systems or miss useful filters. This skill gives Claude, Codex, and Claude Code concise TUnit commands for controlled Playwright execution.

Blazor components can behave incorrectly when static rendering and interactive rendering are mixed. This skill shows how RendererInfo helps adapt UI behavior across render modes.

CSS animations can make floating UI content reject focus before keyboard handlers are ready. This skill guides Claude, Codex, and Claude Code through a retry-based focus pattern and matching tests.

Multi-step delivery work often loses context between validation, review, commits, and CI recovery. This skill gives Claude, Codex, and Claude Code a structured workflow for chaining those steps with evidence checkpoints.

AI assistants can claim work is complete before checking the result. This skill adds a final checkpoint that requires tests, build output, sources, or a clear reason for skipped verification.

OpenTelemetry work often requires precise semantic convention and collector details. This skill guides Claude, Codex, and Claude Code to search bundled OTel documentation and answer with source references.

Complex tasks can produce uncertain or overconfident AI responses. This skill helps assistants identify struggle signals and route difficult decisions to deeper analysis.

MSBuild and central package management drift can be hard to catch during reviews. This skill runs a focused .NET linter and explains violations.

Skill authors need a clear starting point for consistent AI instructions. This template provides standard sections for overview, protocol, examples, red flags, requirements, and related links.

Technical answers often fail when model memory is stale about versions, releases, and dates. This skill adds a verification checkpoint using local assertions and official web sources before the assistant answers.

Teams lose trust when commits hide failures by removing checks or suppressing warnings. This skill adds a Git integrity gate that catches common shortcut patterns before commit.

Complex coding work can miss architecture risks, implementation bugs, and unverified assumptions. This skill coordinates competing reviewer perspectives and merges their findings before deeper analysis.

Code changes can hide security flaws, regressions, and maintainability problems. This skill guides Claude, Codex, or Claude Code through a structured review before commit or merge.

AI assistants can claim work is complete before local tests and remote CI actually pass. This skill creates a strict verification workflow with local test runs, GitHub Actions monitoring, and evidence-based reporting.